HAL Id: hal-00514607
https://hal.archives-ouvertes.fr/hal-00514607
Submitted on 3 Sep 2010
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Development of specific diagnostic test for small ruminant lentivirus genotype E
Ramses Reina, Elena Grego, Margherita Profiti, Idoia Glaria, Patrizia Robino, Antonio Quasso, Beatriz Amorena, Sergio Rosati
To cite this version:
Ramses Reina, Elena Grego, Margherita Profiti, Idoia Glaria, Patrizia Robino, et al.. Development
of specific diagnostic test for small ruminant lentivirus genotype E. Veterinary Microbiology, Elsevier,
2009, 138 (3-4), pp.251. �10.1016/j.vetmic.2009.04.005�. �hal-00514607�
Accepted Manuscript
Title: Development of specific diagnostic test for small ruminant lentivirus genotype E
Authors: Ramses Reina, Elena Grego, Margherita Profiti, Idoia Glaria, Patrizia Robino, Antonio Quasso, Beatriz Amorena, Sergio Rosati
PII: S0378-1135(09)00179-5
DOI: doi:10.1016/j.vetmic.2009.04.005
Reference: VETMIC 4406
To appear in: VETMIC Received date: 19-12-2008 Revised date: 13-3-2009 Accepted date: 3-4-2009
Please cite this article as: Reina, R., Grego, E., Profiti, M., Glaria, I., Robino, P., Quasso, A., Amorena, B., Rosati, S., Development of specific diagnostic test for small ruminant lentivirus genotype E, Veterinary Microbiology (2008), doi:10.1016/j.vetmic.2009.04.005
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Accepted Manuscript
Development of specific diagnostic test for small ruminant lentivirus genotype E 1
Ramses Reina
1, Elena Grego
1, Margherita Profiti
1, Idoia Glaria
2, Patrizia Robino
1, Antonio 2
Quasso
3, Beatriz Amorena
2, Sergio Rosati
1*3 4
1
Dipartimento di Produzioni Animali, Epidemiologia ed Ecologia, Facoltà di Medicina Veterinaria, 5
Università degli Studi di Torino, Via Leonardo da Vinci, 44, 10095 Grugliasco (TO), Italy.
6
2
Instituto de Agrobiotecnología (CSIC-UPNA-Gobierno de Navarra) Ctra. Mutilva Baja s/n 31192 7
Mutilva Baja, Navarra, Spain.
8
3
Dipartimento di Prevenzione, Servizio veterinario, Divisione sanità animale ASL19 Asti, Italy 9
10 11 12
*Corresponding Author 13
Via Leonardo da Vinci 44 14
10095 Grugliasco (TO) 15
tel ++39 011 6709187 16
fax ++39 011 6709196 17
e-mail [email protected] (S. Rosati) 18
19
Manuscript
Accepted Manuscript
Abstract 20
Small ruminant lentivirus (SRLV) belonging to the highly divergent genotype E has recently been 21
identified in the Italian goat breed Roccaverano. In this report we have developed a specific 22
serological test based on recombinant matrix/capsid antigen fusion protein. Performance has been 23
evaluated and compared with a similar test based on genotype B antigen. Herds under study were 24
selected according to the infectious status characterized by blood PCR and sequencing. Results 25
clearly showed that B and E based recombinant ELISA only detected homologous infection and an 26
apparent cross-reactivity was recorded in a herd in which co-infection was present. Three 27
commercially available ELISAs showed different abilities in detecting genotype E infection, being 28
the whole virus-based immunoassay the best choice. Genotype E-recombinant antigen was not 29
detected in ELISA by three commercially available Mabs known to be cross-reactive among CAEV 30
and MVV capsid antigens, further supporting the high divergence of the E genotype from others.
31
Finally, a SRLV-free herd according to commercial ELISA testing, was analysed in the same area 32
where genotype E was identified and few animals belonging to Roccaverano breed were found 33
slightly reactive with the E antigens. Our results suggest that the prevalence of genotype E in other 34
small ruminant populations may be conveniently estimated using a comparative assay based on a 35
combination of genotype specific recombinant antigens and may highlight a wider space in which 36
SRLVs evolve.
37 38
Keywords 39
Small ruminant lentivirus, genotype E, recombinant antigen, diagnosis 40
41
42
43
44
45
Accepted Manuscript
46
Introduction 47
Small ruminant lentiviruses (SRLVs) are a heterogeneous group of viruses affecting sheep and goat 48
and are responsible for chronic debilitating diseases known as Maedi Visna (MV) and Caprine 49
Arthritis-Encefalitis (CAE) (Pepin et al., 1998). Viral isolates characterized so far, show different 50
genetic, antigenic and biological properties and are no longer considered species-specific (Pisoni et 51
al., 2005; Shah et al., 2004b). From an antigenic point of view, most SRLVs can be classified as 52
MVV-like or CAEV-like, corresponding to genotype A and B respectively (Shah et al., 2004a).
53
Previous studies have suggested that early serological diagnosis can be achieved using homologous 54
antigen (Lacerenza et al., 2006). Nevertheless, most of the currently available diagnostic tests are 55
produced using a single strain-based antigen preparation which is believed to detect cross-reacting 56
antibodies against epitopes located in structural proteins (Gogolewski et al., 1985). Recently, a 57
novel genotype E has been identified in the local breed Roccaverano in north-west Italy. First 58
sequences were obtained by chance in a caprine herd, using a set of degenerated primers designed 59
to amplify a gag fragment from the majority of known genotypes, encompassing major linear 60
capsid antigen epitopes. Following a preliminary sequence screening, it seemed quite clear that this 61
viral cluster might have escaped diagnosis in the field using conventional antigen preparations, 62
likely due to the low similarity found in the major immunodominant regions (Grego et al., 2007). A 63
viral strain was subsequently isolated from an apparently healthy goat highly reactive by ELISA 64
against genotype E-derived major capsid antigen epitopes (Reina et al., 2009). Genetic features of 65
this genotype have been described in three epidemiologically unrelated herds. The complete 66
genome (~8,4Kb) presented two major deletions corresponding to the dUTPase subunit of the pol 67
gene and to the vpr accessory gene. Based on previous studies in which such subunits were 68
independently deleted from a pathogenic infectious clone, these deletions could explain why the 69
viral cluster is not related to any known clinical signs, representing a natural, well host-adapted, low 70
pathogenic lentivirus (Harmache, 1996; Turelli, 1996; Zhang, 2003). Moreover, preliminary
71
Accepted Manuscript
epidemiological data suggest that infection can persist in the population through familiar lineage, 72
with a low tendency to spread horizontally. To date, no information is available on the prevalence 73
of genotype E in larger goat populations due to the lack of a specific antibody detection system. To 74
address this issue, in this report we have developed genotype E specific recombinant antigens which 75
were tested with a panel of monoclonal antibodies (Mabs) known to be reactive against CAEV and 76
MVV capsid antigens. These genotype E derived antigens were also used to develop an indirect 77
ELISA in order to test a panel of goat sera belonging to herds in which the SRLV infectious status 78
was determined by PCR product sequence analysis. Serological and sequence data were in 79
agreement highlighting the importance of using genotype specific tests when determining SRLV 80
seroprevalence, assessing SRLV-free status and searchng for epidemiological information.
81 82
Material and Methods 83
Virus and plasmids 84
Roccaverano strain was originally isolated using mammary gland explants from an adult goat and a 85
complete proviral sequence was obtained (Genbank accession number EU293537). The gag gene 86
was amplified by concatenating overlapping PCR fragments and cloned in pCRTopoXL 87
(Invitrogen). The gene fragment coding for Matrix (P16) and major Capsid Antigen (P25) was 88
subsequently amplified and was subcloned between the BamH1/EcoR1 sites of pGEX6His 89
following site-directed PCR-mediated mutagenesis suppressing an internal EcoR1 restriction site.
90
This plasmid, derived from pGEX6P prokaryote expression vector (GE Healthcare), was modified 91
by inserting an in frame 6Xhis-tag between EcoR1 and Sal1 restriction sites thus allowing a double 92
step affinity purification.
93
Expression and purification of P16-25 recombinant antigens and P25-B3 subunit epitopes 94
Transformed E.coli BL21 bacteria were induced at early log phase for 2 h with 0.5mM IPTG under 95
agitation. Bacterial cells were recovered by centrifugation and lysed by physicochemical methods.
96
Recombinant GST/P16-25/6H fusion protein was recovered in the soluble fraction and the first
97
Accepted Manuscript
affinity step was carried out in batch using glutathione Sepharose 4B (GE Healthcare). GST 98
cleavage was achieved in pooled eluted fractions using PreScission Protease (2U/mg) (GE 99
Healthcare). Solution containing GST and P16-25/6H was dialyzed for 24h to remove reducing and 100
chelating agents and loaded into a Hi-Trap chelating HP column (GE Healthcare), positively 101
charged with nickel ions. Following immobilised metal chelate affinity chromatography, purity and 102
yield of recombinant antigen was estimated by SDS-PAGE and DC protein assay (BioRad).
103
Using the same protocol, the same gag antigens derived from strain It-Pi1 (genotype B) were 104
employed to generate the antigenic CAEV-like counterpart.
105
Subunit immunodominant capsid antigen epitopes of MVV and CAEV had previously been 106
characterised. In this study a third version using genotype E (P25-B3) derived sequence was 107
produced, generating a GST fusion protein as previously described (Grego et al., 2002; Rosati et al., 108
1999).
109
Blood samples, polymerase chain reaction and sequencing 110
Twelve caprine herds were selected in this and in a previous study (Grego et al., 2007). Heparinized 111
blood samples were obtained from a number of adult animals, representative of each herd: and 112
DNA was extracted from white blood cells using DNA blood kit (Qiagen). A gag nested PCR 113
previously developed (Grego et al., 2007) and known to detect the highest number of SRLV 114
genotypes/subtypes was applied to each sample and all positive results with suitable bands were 115
sequenced. After, this preliminary screening, five herds of Roccaverano breed were selected: a first 116
herd, BL (n=52) in which only genotype E was detected; a second herd, NG (n=40) in which 117
genotype B and E were detected; and a third group of goats, TM (n=20), BM (n=18) and CF (n=6) 118
in which only genotype B was present. In this study, all caprine herds were retested when possible 119
at completion and additional sequences were obtained.
120
To date, the presence of genotype E had been limited to few herds of the Roccaverano breed.
121
Therefore, an additional three-breed long term SRLV negative herd (n=400) was also included for
122
Accepted Manuscript
serological testing. The herd consisted of 109 Roccaverano, 107 Saanen and 184 French Alpine 123
goats.
124 125
Serum samples and ELISAs 126
For P16-25 recombinant ELISA, microplates (Immunomaxi TPP) were coated with 100 ng of P16- 127
25 derived from B and E genotypes or water as negative antigen (Fig. 1). Plates were allowed to dry 128
overnight at 37°C and then blocked with 2.5% bovine casein for 1h at 37°C. After four washes, 129
serum samples were diluted 1/20 in phosphate-buffered saline containing 1.25% casein and 130
incubated for 1h at 37°C. Subsequently to the washing step, anti-sheep/goat IgG peroxidase labelled 131
Mab diluted in the same buffer was added and the plates incubated as above. After a final washing 132
step, the reaction was developed with ABTS and plates were read at 405 nm. Net absorbances were 133
obtained by subtracting the absorbance of negative antigen from the absorbance of each 134
recombinant antigen and visualized by box plots. Cut off was previously defined for genotype B as 135
having a reactivity of >40% relative to the positive control serum reactivity included in each plate 136
(Lacerenza et al., 2006). The same absorbance value was applied for genotype E, due to the lack of 137
a truly negative flock regarding this genotype.
138
For subunit ELISA (P25-B3 epitope), microplates were coated with 220 ng of GST-B3 derived 139
from genotype B (sequence KLNEEAERWRRNNPPPP), genotype E (sequence 140
KLNKEAETWMRQNPQPP) and an equimolar amount of GST as negative control. Net 141
absorbances were obtained by subtracting the GST antigen absorbance from that of each 142
recombinant subunit.
143
Three commercially available ELISAs, based on whole virus (brand A), double recombinants 144
(brand B) or recombinant and synthetic (brand C) antigens, were used to detect and quantify the 145
infection status in BL herd. Assays were carried out according to the manufacturer’s protocols.
146
Three commercially available monoclonal antibodies (Mabs), namely 5A1, 10A1 and 8B1, known 147
to be reactive against CAEV and MVV capsid antigen (McGuire et al., 1987) were obtained from
148
Accepted Manuscript
VMRD, Inc. (Pullman WA, USA) and tested in P16-25 recombinant ELISA. Mabs were used at 149
dilution of 1μg/well, using mouse-specific secondary antibody. In the latter experiment, a 150
previously produced recombinant capsid antigen from the MVV strain K1514 (Rosati et al., 1999), 151
was used to confirm Mabs cross-reactivity between genotype A and B.
152 153
Western blot 154
The genotype B and E recombinant P16-25 proteins were also tested by Western blot using the 155
same three Mabs (5A1, 10A1 and 8B1), as well as a polyclonal serum from mice immunized with 156
recombinant P16-25 of genotype E (positive control) and serum from the same goat from which the 157
Roccaverano strain (genotype E prototype) had originally been isolated.
158
Sequence analysis and phylogenetic trees 159
In order to create phylogenetic trees, a model of molecular evolution was estimated using a 160
hierarchical likelihood ratio test approach and the Akaike information criterion (Akaike, 1973) 161
implemented in the software ModelTest ver. 3.7 (Posada and Crandall, 1998, , 2001). Bayesian 162
methods implemented in the computer program MrBayes ver. 3.1.1 (Huelsenbeck and Ronquist, 163
2001. Ronquist and Huelsenbeck, 2003) were used to create phylogenetic trees and to assess 164
statistical support for clades. A Markov chain Monte Carlo search for 1,000,000 generations using 165
two runs with four chains (temperature = 0.05) was performed and results were represented as a 166
50% majority rule consensus tree. Tree statistics and phylogenetic manipulations were calculated 167
from the computer program PAUP* ver. 4.0b10 (Swofford, 2000). Genetic diversity was expressed 168
as nucleotide diversity (Nei, 1987), or the mean proportion of nucleotide differences among 169
sequences.
170
Statistical analysis 171
Maximum expected prevalence of both genotypes in the herds was estimated using WinEpiscope 172
2.0 (http://www.clive.ed.ac.uk/winepiscope/).Agreement beyond chance between genotype E
173
Accepted Manuscript
specific ELISA and commercial assays was analysed using Fleiss’κ coefficient (Fleiss, 1981).
174
Differences in O.D. values between tests were evaluated using Wilcoxon test.
175
Nucleotide sequence accession numbers 176
Nucleotide sequences reported in this paper were deposited in the GenBank database with accession 177
numbers FJ547242-46. A subset of additional sequences (EF675997-6026; EU726488-525) had 178
been generated in previous studies (Grego et al., 2007; Reina et al., 2009).
179
Results 180
Recombinant antigen derived from genotype E was successfully expressed and purified. SDS- 181
PAGE of the purified product revealed a single protein band of a molecular weight corresponding to 182
the expected size of matrix and capsid antigen fusion protein (41kDa) (Fig 1).
183
Out of 124 blood DNA samples, 56 were PCR positive and the 39 suitable for sequence analysis, 184
were used to construct a phylogenetic tree using CAEV-CO (M33677), Roccaverano (EU293537) 185
and K1514 (M10608) as reference strains. As shown in Fig. 2, all sequences, representative of herd 186
BL belonged to genotype E (8 sequences out of 12 positive PCR reactions). This sample size 187
allowed us to estimate a maximum expected prevalence of B genotype lower than 25% (CI 95%) 188
within this herd.
189
Conversely, sequences obtained from herds TM, BM and CF were grouped into the B1 subtype (15 190
sequences out of 22 PCR reactions), allowing an estimation of a maximum expected prevalence of 191
E genotype lower than 13.5% (CI 95%). Finally, both infections were detected in herd NG with a 192
high degree of co-infections (B and E genotypes) within the same animal, as a previously described 193
genotype specific PCR had revealed (Grego et al., 2007).
194
Box plots of serum reactivity against B and E recombinant antigens are shown in Fig 3. In herd BL 195
and the TM-BM-CF herd group, only one homologous antigen was clearly detected among infected 196
animals based on infectious status determined by phylogenetic analysis, while an apparent cross- 197
reactivity between the two antigens was present only in herd NG. To further study the serological 198
reactivity against genotype specific gag antigens, sera were also tested by subunit ELISA, using
199
Accepted Manuscript
P25-B3 epitope from genotypes B and E. Indeed, several samples reacted in a type specific manner 200
(Fig. 4). Serum samples belonging to herd BL showed a distribution mainly oriented towards the E 201
derived epitope and on the contrary, B1 infected herds (TM-BM-CF) were clearly reactive only 202
with the B derived epitope (Fig 4 a and 4b). Herd NG showed a wider distribution of absorbances 203
reaching high values against both peptides (Fig 4 c).
204
Genotype E derived p16-25 antigen reacted also with few sera from the long term negative herd 205
tested and these sera were from Roccaverano goats, while no reactivity was recorded in Saanen and 206
French Alpine goats cohabiting in the same herd (not shown).
207
Commercial ELISAs showed a different capacity in detecting infection in herd BL, being the whole 208
virus ELISA in perfect agreement with genotype E derived ELISA (κ= 1), followed by the double 209
recombinant ELISA (κ= 0.83) and the recombinant/synthetic ELISA (κ= 0.47 ).
210
Distribution of absorbance values in infected animals of the same herd revealed that the best signal - 211
to - noise ratio belongs to genotype E antigen and is significantly higher than those obtained from 212
the three commercial ELISAs (p<0.05) (Fig.5).
213
None of the three Mabs showed reactivity against recombinant antigen derived from genotype E in 214
ELISA while antibody cross-reaction was confirmed when using CAEV and MVV capsid antigens 215
(Fig. 6a). Western blot analysis (Fig. 6b) showed a strong reaction when using p16-25 from 216
genotype B and a weak signal against p16-25 of genotype E, likely due to intrinsic higher 217
sensitivity of WB compared to ELISA under the test conditions applied.
218 219
Discussion 220
In this study we demonstrate that the highly divergent SRLV genotype E can be detected in a type 221
specific manner by a comparative assay using gag derived antigens from different genotypes. The 222
true prevalence of viral E genotype might have been underestimated so far for two reasons: i) 223
commercial ELISAs may detect infection without yielding specific serotype information; and ii) 224
some animals may escape diagnosis based on immunodomiant epitopes highly divergent from E
225
Accepted Manuscript
genotype antigens. The development of a first serological tool specific for genotype E has become 226
therefore essential to monitor natural infection by this genotype in the field. We chose gag derived 227
structural proteins since it is well documented that anti-gag derived antibody response remains 228
detectable for a long time, if not life long (de Andres et al., 2005), and cross-reactivity between 229
MVV and CAEV P16-25 is acceptable except in the early stage of infection when homologous 230
antigen is more quickly recognised (Lacerenza et al., 2006). Lack of sensitivity of CAEV P16-25 in 231
detecting genotype E infection and vice versa was not surprising. All the diagnostically relevant 232
epitopes characterized so far in the gag encoded proteins (summarized in Table 1), show 233
divergences between genotype E and CAEV or MVV. Commercial ELISA based on whole virus 234
preparations demonstrate that genotype E can be detected, likely due to the presence of a larger 235
panel of cross reacting epitopes which may compensate the high variability between genotype E and 236
ELISA antigen. The other commercial ELISAs used are based on a more restricted antigen display 237
and we cannot exclude that, for example, the TM epitope of genotype E might elicit cross-reacting 238
antibodies recognised by heterologous antigen preparations.
239
We previously described P25 immunodominant epitope and matrix protein as specifically detecting 240
A or B genotype infections (Rosati et al., 1999; Grego et al., 2005). Here, we show that genotype E 241
derived P25 immnodominant epitope also recognises only homologous infection. Lack of reactivity 242
of three Mabs, known to be reactive against CAEV and MVV, with genotype E P25 protein was 243
surprising and demonstrates that there are at least three different type specific epitopes in P25. This 244
is relevant when aiming for antigenic detection of a wide spectrum of SRLVs. The hypothesis that 245
genotype E infection could have escaped through traditional diagnostic pressure (test and slaughter 246
policy), was evaluated in a SRLV free herd in which few old aged animals of the Roccaverano 247
breed were slightly reactive against genotype E recombinant antigen. We cannot rule out a false 248
positive reaction since an attempt to amplify viral sequences from the same animals gave 249
inconclusive results. It should be taken into account that no positive reaction was recorded in any of 250
the 292 animals belonging to other breeds of the same herd. On the other hand, animals with a low
251
Accepted Manuscript
proviral load could be misdiagnosed using PCR. Interestingly, young animals of the three breeds 252
(Roccaverano, Saanen and French Alpine) were negative, confirming that horizontal transmission is 253
probably hampered due to genetic deletions present in this low pathogenic virus. Thus, artificial 254
feeding of kids, a prevention strategy for SRLV control still used in the herd, is an efficient way to 255
co-eradicate both CAEV and genotype E infections.
256
To date we could identify only three herds infected with genotype E and no more than about a 257
hundred E-infected animals, all belonging to the Roccaverano goat breed (total heads 1500) which 258
was at risk of extinction in the seventies. We are therefore aware that the geographical distribution 259
of genotype E may have a very limited impact in other countries. However, this study shows that 260
the lack of E specific serological tests does not allow the determination of the true prevalence of 261
genotype E infection. Molecular tools such as PCR are now available for genotype E as well as a 262
congruous number of GenBank deposited sequences. The low agreement between PCR and ELISA 263
in detecting genotype E infected animals was not surprising, since this has been widely 264
demonstrated by other authors regarding A or B genotypes. This is likely due to low viral load or to 265
high heterogeneity at the primer binding site (de Andres et al., 2005). It should also be noted that 266
not all the recently described PCRs developed for generic detection of SRLVs may be used for 267
genotype E identification. Primers described by Shah (Shah et al., 2004a) designed to amplify 1.2kb 268
of pol gene, will result in a product of 0.8kb, resembling unspecific, due to internal dUTPase 269
deletion. In addition, due to the same deletion, PCR primers described by Eltahir (Eltahir et al., 270
2006) will fail to amplify as a result of the lack of forward primer binding site.
271
In conclusion, our results demonstrate that, despite the limited information on the distribution of 272
genotype E in small ruminant population, specific antigen design is required for accurate diagnosis.
273
Subunit ELISA using a panel of A, B and E immunodominant region of capsid antigens can be used 274
as a rapid preliminary screening test that would allow serotype determination. On the other hand, 275
evidence regarding this highly divergent genotype has widened the space in which SRLVs have 276
evolved with potential implications for control strategies.
277
Accepted Manuscript
Acknowledgement 278
This work was supported by Regione Piemonte: Ricerca Scientifica Applicata 2008 and Italian 279
MIUR.
280
281
282
283
284
Accepted Manuscript
Fig legends 285
Fig 1 286
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis showing expression and purification of 287
recombinant antigens in E.coli. MW, molecular-weight standard; lane 1 E. coli lysate; lanes 2 and 4 288
bacterial lysate expressing GST/P16-25 of It-Pi1 (genotype B) and Roccaverano (genotype E) 289
respectively; lanes 3 and 5, affinity purified recombinant P16-25 of the same strains after GST 290
cleavage and IMAC.
291
Fig 2 292
Phylogenetic tree of amplified gag region from different herds. The reference strains of genotype A 293
(M10608), B (M33677) and E (EU293537) are shown.
294
Fig 3 295
Box plots of net absorbances in ELISA testing of sera from BL herd (panel A), NG (panel B) and 296
TM-BM-CF group (panel C) against recombinant P16-25 of genotypes B (left box) and E (right 297
box). Bottom and top of each box represent the first and the third quartiles of the distributions. The 298
median is represented by the line inside the box.
299
Fig4 300
Net absorbance against P25-B3 subunit ELISA: each serum sample was tested against the B derived 301
epitope (X axis) and E derived epitope (Y axis). Samples are from BL herd (panel A), NG (panel 302
B) and TM-BM-CF group (panel C).
303
Fig 5 304
Distribution of O.D. values in BL herd using different serological tests. Sera which were positive to 305
at least two ELISA tests were considered as true positive. Brand A: whole virus based ELISA (O.D.
306
450nm); brand B: double recombinant based ELISA (O.D. 450nm-595nm); brand C recombinant 307
and synthetic based ELISA (O.D. 450nm-595nm); genotype E specific recombinant P16-P25 308
ELISA (O.D. 405nm). Bottom and top of each box represent the first and the third quartiles of the 309
distributions repectively. The median is represented by the line inside the box.
310
Accepted Manuscript
Fig.6 311
a) Monoclonal antibody reactivity in ELISA (O.D. 405 nm) against capsid antigen of CAEV isolate 312
(McGuire et al. 1987). Net absorbances against recombinant P16-25 of genotypes E and B and 313
recombinant P25 of genotype A are shown.
314
b) Western blot analysis of the same Mabs (5A1, 10A1 and 8B1) against P16-25 of genotypes B 315
(lanes 1, 2 and 3) and E (lanes 4, 5 and 6). Lanes 7 and 8 were developed using a polyclonal mouse 316
antiserum raised against E derived fusion protein, and a serum from an E infected goat respectively.
317
MW, molecular weight marker.
318 319
Table: Immunodominant linear epitopes of SRLV identified so far along the structural proteins in 320
genotypes A, B and E. Genbank accession numbers, source references and amino acid sequences 321
from each genotype are indicated. Dashes represent the same amino acid as that of genotype B.
322 323 324
325
326
327
328
329
330
331
332
333
334
335
336
Accepted Manuscript
Genotype GenBank
Matrix (Grego et al., 2005)
Capsid antigen B (Grego et al., 2002)
Capsid antigen D (Rosati et al., 1999)
Nucleoprotein (Lacerenza et al., 2008)
Transmembrane (Bertoni et al., 1994) B M33677 KLLTPEESNKKDFMSL LNEEAERWRRNNPPPPA VQQASVEEKMQACRDVGSE GNGRRGIRVVPSAPPME ELDCWHYHQYCITS A M10608 -N---TS-RE-A-- --D---V-Q---G-N ----T--- --N---P---L ---QH-—V-- E EU293537 RSM----ESR---V-- --K---T-M-Q—Q—-GP A-TST----L----E-E-S --SQ--P---Q -I---G—-V--
